When an isometrically activated muscle is stretched or shortened the isometric steady-state force after the length change is increased (residual force enhancement) or decreased (force depression), respectively compared to a purely isometric contraction. This behavior has been observed consistently from the single sarcomere to the whole muscle level. However, the results for voluntary contractions in vivo are controversial and there are no studies for maximal voluntary contractions of medium sized muscles like the human ankle dorsiflexors. We investigated the effect of active muscle stretching and shortening for in vivo human tibialis anterior (n = 12) for maximal voluntary contractions for two magnitudes of stretching (15° and 30°) and two speeds of contraction (10°/s and 45°/s). Torques during stretches were higher compared to the purely isometric reference contractions and peak torques occurred prior to the end of the stretch. During the stretch, muscular activity decreased after peak torque had been reached for the high speed stretch experiments. In the steady-state phase following stretch, torque was increased for all experimental conditions but not for all time periods following stretch. The amount of residual force enhancement was independent of amplitude and speed of stretch. In the steady-state phase following shortening, torques were decreased compared to the isometric reference contractions and force depression was increased with increasing speeds of shortening.